Transcranial Magnetic Stimulation: A Review - IEEE Xplore

Transcranial Magnetic Stimulation: A Review BY PAOLO MARIA ROSSINI, LUCA ROSSINI, AND FLORINDA FERRERI

I

n recent years, noninvasive brain stimulation methods have been proposed as the next-generation technology to probe and eventually interfere with brain function. Transcranial magnetic stimulation (TMS) is a 20-year-old technique originally introduced to investigate nervous propagation along the corticospinal tract, spinal roots, and peripheral nerves in humans. TMS is extensively used in clinical neurophysiology, including rehabilitation and intraoperative monitoring. TMS is a noninvasive and painless technique that allows researchers to stimulate discrete brain areas. This technique involves delivering a brief magnetic pulse to the scalp through a coil; the stimulating effect depends on the geometry of the stimulating coil with respect to the head and of the waveform of the current pulse flowing through the coil. With the commonly used stimulation intensities and coils producing relatively focal types of stimuli, the cortex is activated within an area of a few square centimeters. The technique of stimulating the motor cortex and recording the muscle twitch or surface potential responses in the periphery was established by the work of Merton and Morton [1], who showed in 1980 that it was possible to stimulate the motor cortex of the human brain through the intact scalp by using short-duration and large-amplitude electrical pulses delivered through a pair of surface electrodes [transcranial

Digital Object Identifier 10.1109/MEMB.2009.935474

84

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

electrical stimulation (TES)]. After stimulation of the corticospinal tract, a motor twitch named motor-evoked potential (MEP) was recorded from the connected muscles in an awake and collaborative subject. The required stimulation intensity, however, was uncomfortable and poorly tolerated [2]. The major problem with this form of stimulation is in fact that only a small fraction of the applied current actually flows into the brain. More electric current flows between the electrodes on the scalp and produces local discomfort and painful contraction of the scalp muscles. In the mid-1980s, however, Barker et al. demonstrated for the first time that the stimulation of the human motor cortex and peripheral nerves © STOCKBYTE can be performed using a brief and strong external magnetic field applied through a wire coil that is able to produce muscle twitches that can be recorded by surface electromyography (EMG) without causing distress or pain. The magnetic pulses stimulated neural tissue through the cortically induced electric field, depolarizing cell membranes [Figure 1(a)], [3], [4]. It is important to acknowledge that the actual pathways being investigated are not known; however, they incorporate the fastest conducting fibers that certainly include the pyramidal tracts [5]–[7]. When applied on the scalp overlying the primary motor cortex (M1), single-pulse TMS allows routine evaluations of the excitability and conductivity of corticospinal motor pathways. This approach has been largely applied in investigating movement physiology in healthy patients, patients with neurological disorders, and postlesional follow-up studies of patients with plastic

0739-5175/10/$26.00©2010IEEE

JANUARY/FEBRUARY 2010

Control Size (%)

cerebral reorganization. Furthermore, paired-pulse TMS of stimuli separated by a variable interval (paired-pulse TMS), allows the investigation of intracortical mechanisms of inhibior in trains (rTMS). Single-pulse TMS can be used for maption and facilitation. In the late 1990s, a new generation of ping motor cortical outputs, studying central motor conducmagnetic stimulators was introduced, which is able to deliver tion time, and studying causal chronometry in brain–behavior rhythmic trains of several stimuli per second and therefore relations. In paired pulse techniques, TMS stimulation can be called repetitive TMS (rTMS). The terms slow-frequency delivered to a single cortical target using the same coil or to rTMS for rates less than 1 Hz and high frequency for those two different brain regions using two different coils. Paired more than 1 Hz (up to 60 Hz) are generally used. The cutoff at pulse techniques can provide measures of intracortical facili1 Hz is not entirely arbitrary and empirical, because there is tation and inhibition as well as study corticocortical interacevidence that slow- and high-frequency stimulations produce tions. In paired pulse protocols, two stimuli are released: the relatively distinct effects, both on direct measures of brain first (S1) or conditioning and the second or test stimulus (S2). activity and behavior, via psychophysical techniques. CogniInhibition occurs by setting S1 below the threshold and S2 tive neuroscientists therefore have the opportunity to interfere well above the threshold and by using extremely brief interstitransiently with the behavior in several domains (i.e., motor, mulus intervals (ISIs) in the order of 1–3 ms, while facilitavisual, sensory, memory, language; see [8]). Such an approach tion—with the same conditioning/test intensities—can be can be used not only for a better understanding of cortical obtained by setting ISI in the order of 7–12 ms [Figure 1(b)]. functions (namely, to reveal the causal relationship between Plastic modifications of the intracortical excitability can be the targeted area and the task under study) but also in attempts achieved by pairing a peripheral stimulus to a single TMS to modulate a given brain function in healthy people and those stimulus, the so-called paired associative stimulation (PAS). with pathology, thereby extending and somewhat challenging PAS induces plastic cortical changes in sensorimotor cortex the concept of the virtual lesion patient previously introduced by coupling a single electric stimulus, delivered on a periphto depict the interference effects of rTMS. The existing literature eral nerve, with a single magnetic pulse delivered on the scalp. indicates that these effects are generally short lived. However, The direction of cortical plasticity induced by paired the possibility of interfering with complex cognitive funcConditioning TMS tions in the long term—espeMagnetic Field 1 cially when several sessions of Controll Test TMS rTMS are repeated during fol2 low-up sessions—opens new Interstimulus strategies for modifying brain– Interval al Coil = 2 ms s behavior relationships, making TMS a hypothetical rehabilitative tool. Currently, despite Scalp = 12 ms x fascinating attempts to use Corte TMS in this way, large-scale Electric Field 0.5 mV applications in clinical settings 25 ms are still lacking. The aim of this 400 Pyramidal Neuronal Axons review, therefore, is to provide the reader with a brief overview 300 of the TMS background (until + + + − Depolarization now, more than 5,000 scientific − − − + −− papers have been published on − − + +− + 200 this technique) for a better + + − −+ − + Electric Field − − understanding of the current + − −+ 100 + − and future uses of TMS. + − + − − + Afferent volleys to the motor + − cortex from peripheral nerves 0 Membrane of the Axon + 15 20 0 5 10 significantly modulate the cortInterstimulus Interval (ms) ical excitability and the final (a) (b) output down to spinal motoneurons whenever an appropriate interval is separating the Fig. 1. (a) The electric current flowing in the TMS coil(s) induces a magnetic field, which penarrival of the sensory input etrates the layers of tissues separating the brain from the outside (scalp). The magnetic field from the impact of the transcra- induces an orthogonal electric field in the underneath brain cortex, which causes the movenial stimulation on the motor ment of ions. Neuron axons orthogonal to the electric field are maximally perturbed in their membrane potential and undergo axon depolarization or hyperpolarization depending on cortex [9]–[12] (Figure 2). Within this historical re- axon orientation with respect to the current flow direction. In the case of depolarization, if construction, it is worth re- the electric potential between the outside and the inside of the axon goes below the actiminding the reader of some vation threshold, the neuron responds by firing an action potential. (b) Examples of recovery nomenclature. TMS can be curves of MEP amplitudes during paired-pulse TMS through the same coil delivered to the applied one stimulus at a time right motor cortex; note that with 2 and 12 ms ISI there is a significant inhibition and facilita(single-pulse TMS), in pairs tion respect to the control condition of MEPs elicited by test stimuli alone.



85

The direction of cortical plasticity induced by paired stimulation depends on the interval between the stimuli.

stimulation depends on the interval between the stimuli. If sensory input from median nerve stimulation reaches the motor cortex at appropriate intervals prior to magnetic stimulation, PAS potentiates local cortical excitability. When multiple stimuli of TMS are delivered in trains, one can differentiate conventional and patterned protocols of repetitive stimulation. For conventional protocols, there is an agreement on the term rTMS. Patterned rTMS refers to repetitive application of short rTMS bursts at a high inner frequency interleaved by short pauses of no stimulation. Most used to

Flexor Pollicis Brevis

1 mV 10 ms 10 12 16 18 20 22 24 26 28 30 32 34 38 42

date are the different theta burst (TBS) protocols in which short bursts of 50-Hz rTMS are repeated at a rate in the theta range (5 Hz) as a continuous (cTBS) or intermittent (iTBS) train [13], [14]. Lasting inhibitory aftereffects of 1-Hz rTMS and cTBS and facilitatory aftereffects following high-frequency rTMS and iTBS were found, which can be explained by various mechanisms, including synaptic changes resembling experimental long-term depression (LTD) and long-term potentiation (LTP), as well as shifts in network excitability, activation of feedback loops, and activity-dependent metaplasticity [15], [16]. Regarding patterned rTMS, most TBS protocols employed to date replicate the original ones explored by [13]: for N20 cTBS, three pulses at 50 Hz are applied at 5 Hz for 20 s 17.3 (300 total stimuli) or 40 s (600 stimuli). For iTBS, 20 2-s periN30 ods of cTBS, each separated 2uV from the following by 8 s, are 10ms 27.8 applied. Recently, quadripulse stimulation (QPS) [17] has been added to patterned rTMS promV cedures to induce long-term changes of cortical excitabil3 ity. Repeated trains of four monophasic pulses separated 2 by ISIs of 1.5–1,250 ms 3sd produced facilitation (at short intervals) or inhibition (at 1 longer intervals), probably through a modulatory action on intracortical excitatory cir10 14 18 22 26 30 34 38 42 46 cuitry [17].

O = Nonconditional MEP 46

Nerve TCS I.S.I.ms

Basic Principles (a)

(b)

Fig. 2. (a) The top traces are unconditioned MEPs elicited by test TMS alone. The following traces are MEPs conditioned by a conditioning stimulation on median nerve at wrist subthreshold for eliciting a muscle response but well perceived by the subject. Each trace is the superimposition of several responses in each condition. Numbers indicate intervals in milliseconds separating conditioning stimulation of median nerve from test TMS on the scalp. (b) Plots of MEP amplitudes at various intervals. It is clear that an initial dip of inhibition between 10 and 24 ms followed a sharp peak of facilitation, reaching the maximum at about 30 ms. This is about the latency of the frontal component of negative polarity when the median nerve is stimulated at wrist, and somatosensory-evoked potentials are recorded (used with permission from [9]).

86


The physical basis of TMS is described by Maxwell’s equations. Because the electromagnetic fields associated with TMS are of low frequency, the quasi-static approximation of the equations [18] can be applied to the computation of the tissue-induced fields and currents. A time-varying current pulse in the stimulation coil produces a magnetic field


Magnetic stimulation is less affected by distance than electrical stimulation.

according to the Biot–Savart law. The time-varying magnetic field, in turn, induces an electric field according to Faraday’s law. The induced electric field moves charges in the direction of its field lines. If the coil is parallel to the surface of the conductor (head), no surface charges appear due to induction, and the computation of the electric field inside the conductor is simple [19]. Otherwise, charges accumulate at the conductor surface as well as at the interfaces between tissues with different conductivity, generating a secondary electric field. The intensity of the magnetic field can be represented by flux lines around the coil and is measured in tesla (T). The magnetic field is oriented perpendicular to the coil and, for currently available devices, can reach values of up to 4 T even if average values of maximal stimulating intensities around 2 T are currently provided by commercially available devices [3]. E¼

dA DV: ds

The expression for the total induced electric field E inside a conductor has, thus, a term due to induction, represented by vector potential A, and a term from surface charges, represented by scalar potential V [20]. Because the total induced electric field is strongest at the boundaries of any homogenous conductor compartment [21], the stimulating effect of TMS in the brain is concentrated at the cortical surface. The electric field induced in tissue causes the cell membranes to either depolarize or hyperpolarize. If the depolarization of the membrane overcomes its threshold, an action potential is generated (Figure 1). The classical assumption is that the activation of pyramidal neurons by TMS occurs predominantly via interneurons in superficial cortical layers [6], [22]. Macroscopically, the locus of activation in the brain seems to be where the induced field is maximal [23]. Focal activation is achieved by using a figure-of-eight coil or a double-cone coil with two loops, in which the current flows in opposite directions. The induced electric field peaks at the intersection of the coil windings (virtual stimulating cathode) (Figure 3). The stimulating field experienced by a neuron has its duration equal to the first phase of the dB/dt waveform. A greater amount of stored energy is required for longer-duration stimuli to achieve the same change in transmembrane voltage [24]. Therefore, short pulses with rise times less than 100 ls are usually applied.

short time, a magnetic stimulator works by charging one or more energy storage capacitors and then rapidly transferring this stored energy from the capacitor(s) to the stimulating coil as it discharges. In this way, a very high electrical current with a peak value of more than 5,000 A flows from the capacitor(s) through the stimulating coil generating the required magnetic field. The difficulty in producing magnetic nerve stimulators is related to the high discharge currents, voltages, and power levels involved in producing the brief magnetic pulse. Typically, 500 J of energy has to be transferred from the energy storage capacitor into the stimulating coil in around 100 micros. As power (measured in watts) is equivalent to joules per second, the power output of a typical magnetic stimulator during the discharge phase is 5 MW (5,000,000 W). Such an electrical power could provide the electricity used by 1,000 average houses in just 1 ms. During the discharge, the energy that was initially stored in the capacitor in the form of electrostatic charge is suddenly converted into magnetic energy in the stimulating coil in approximately 100 ls. This rapid rate of energy transfer produces a time-varying magnetic field buildup that induces tissue currents in the proximity of the coil in the order of 1–20 mA/cm2. However, the amount of thermal energy delivered to the surrounding tissues due to magnetic stimulation is very small. At maximal output, assuming a maximal stimulus repetition rate of one pulse every 3 s, the average power deposited in the brain is calculated to be less than 2 mW. This represents less than 0.01% of the heat generated in the adult brain because of the normal basal metabolism. Heating of the brain is of the order of 106 °C/pulse and unlikely to cause deleterious effects [24]. Similarly,

Technical Requirements

When magnetic stimulation is performed on the motor cortex, MEPs can be recorded in contralateral, connected muscles. However, large pulses of magnetic field need to be generated to induce intracortical electric fields of sufficient amplitude and duration to provoke the concurrent activation of the surrounding neural tissue. To release enough energy in a very


Fig. 3. Magnetic field distribution over the scalp induced by a figure-of-eight coil (also known as double-cone coil). When the coil is placed parallel to the scalp as shown in the figure, the achieved activation is focal, with a field peak occurring at the intersection of the coil wings.


87

Threshold in human adults is almost independent of age, gender, and hemisphere but varies with different target muscles.

international standards suggest that the continuous thermal energy deposited in tissue by electromagnetic radiation should not exceed 0.4 W/kg to avoid thermal stress. For a typical human adult brain mass of 1.5 kg, this limit is 300 times greater than the average thermal energy deposited by the magnetic stimulator at maximal output even if stimulation is carried out continuously [4]. The stimulating coil, normally housed in a molded plastic package, consists of one or more tightly wound and well-insulated copper coils together with other electronic circuitry such as temperature sensors and safety switches. At the present time, most commercial magnetic stimulators are supplied with a circular coil of 5–10 cm diameter. Different coil types are nowadays available, each with their own advantages and disadvantages. Large coils cannot produce very focal stimulation but can penetrate relatively deep in the brain. Although the circular coil is very useful as a general purpose coil, it does not provide a very defined site of stimulation. For example, with a standard round coil, the induced current in the brain flows in an annulus, underneath the coil, which is usually some 8–12 cm in diameter. Clearly, a large volume of neural tissue may be activated by such a device. Increases in the focality of stimulation can be achieved by tilting these circular coils so that their plane is not tangential to the skull at that point. The greater the angle between the skull and the coil, the more focal is the stimulation. Unfortunately, increased focality of stimulation is offset by a decrease in its effectiveness due to power dissipation in the air by that part of the coil that loses contact with the scalp. Figure-of-eight coils (also termed butterfly or

Single Circular Coil

Double Circular Coil RJ

(a)

(b)

Fig. 4. (a) Circular coil with the produced stimulating field. Notice that the part of maximal intensity is localized along the circumference, while the one with minimal intensity is located right in the center of the coil. (b) Figure-of-eight coil. Notice that the intensity of maximal stimulation corresponds to the point of conjunction between the two wings.

88


double coil) induce an electric field under the junction region of the eight, which is twice as large as that under the two wings [25]. These coils can sustain larger currents because of the lower induction with respect to circular coils with the same number of rings (Figure 4). However, the hypothesis that the double coil stimulates only under its center should be taken with some caution: together with the central peak, there can be peripheral peaks (usually of approximately half the amplitude of the central one) on either side of the winding. Difference with Electrical Stimulation

Like magnetic stimulation, electric stimulation of the motor cortex evokes EMG responses in contralateral, particularly, distal muscles. MEPs resulting from either transcranial electric or magnetic motor cortex stimulation may be used to demonstrate the functional integrity and conduction properties of the central /peripheral nervous motor fibers [1], [26]. In this respect, magnetic stimulation has three main advantages on conventional electrical stimulation. First, the primary benefit of magnetic stimulation is its ability to penetrate all body structures without attenuation, allowing researchers to stimulate regions well below layers of bone, for instance, the brain tissue under the skull. The mechanism of stimulation at the neural level is thought to be the same for both magnetic and electrical stimulation, namely current passes across a nerve membrane and into the axon, resulting in depolarization and the initiation of an action potential that propagates following the well-known rules of nerve conduction [4]. Although the magnetic and its induced electric field should be theoretically unaffected by the spinal vertebrae, when the magnetic coil is placed over the spine there is a stimulation of the nerve roots at their spinal exit (though without stimulation of the spinal cord). It is suggested that the bony structure surrounding the spinal cord interferes with the spread of magnetically induced eddy currents to the spinal cord. The current induced by the magnetic coil is theoretically maximum in the annulus under the coil. The charge builds up on the bone at the points where the induced current loops at the intervertebral foramina, where the spinal roots enter and leave the spine. Therefore, the bony vertebrae act as an insulator between the spinal cord and the external tissue and the current probably tends to flow around the spinal cord rather than through it. Hence, stronger magnetic fields are required for spinal stimulation, and novel coil geometries may be able to improve the coupling between the induced currents and the anatomy of the spine. Second, magnetic stimulation is less affected by distance than electrical stimulation. That is, the electrical field induced with a coil (of 100 mm in diameter) decreases significantly less with increasing distance into the body than with a field induced by currents applied via surface electrodes [4]. Electrical stimulation works by injecting electrical current into the body via


surface, needle, or implanted electrodes. The charge from an electrical stimulator is carried by electrons flowing in the wires to the stimulating electrodes and is transferred to an ionic flow at the electrode–tissue interface. Only a small percentage of these ions actually flows into nearby axons, resulting in membrane depolarization. Magnetic stimulation differs from electrical stimulation, since it uses a pulse of magnetic field to establish an electric field in the tissues, which ultimately results in stimulation. At the stimulating frequencies used with TMS, the magnetic field is not affected by the electrical properties of the body and passes through both bone and soft tissue (and even clothing and air) without being affected by them and without causing large electrical fields at the surface. This means that almost all of the electric field induced by the magnetic stimulation results in membrane depolarization. Also, the field distribution produced by the two stimulation techniques differs substantially. In the electric one, the current flows beneath the electrodes in all directions away from the anode, both radially and tangentially to the cortical surface, stimulating structures close to the surface in this orientation [5]. The magnetic stimulation, instead, produces an electric field that is much more homogeneous and parallel to the surface of the coil, stimulating structures with a different orientation [27]. The absence of current leaks together with its peculiar induced electric field properties were thought to explain the ability of the TMS to stimulate, without pain or any discomfort, deep structures such as the lumbar roots, the brachial plexus, and the sciatic, radial, and femoral nerves in humans. However, the depth of penetration depends on anatomical factors, coil size, coil geometry, and the intensity of the applied stimulus. In the simple homogeneous model, the volume within which straight nerves can be stimulated, for both circular and figure-of-eight coils, is shaped roughly like an egg and has maximum dimensions at the surface, which decreases in cross-sectional area until zero at the maximum depth at which the stimulation threshold is reached. Third, magnetic stimulation does not require either physical or electrical contact with the body. Hence, neither skin preparations are required nor clothing needs to be removed. Although the coil is normally placed in contact with the body, this is just for the operators’ convenience, since stimulation can be perfectly achieved with the coil held some millimeters away from the body. This is extremely valuable in situations where the contact would be otherwise an issue, causing further damage or infection, such as in the stimulation of traumatized regions. The stimulating coil can be freely moved over the area of interest, which makes it rapid and easy to locate the optimal stimulation site [4]. Ever since the introduction of TMS, there has been a considerable debate over which structures within the cerebral cortex are activated. The first hypothesis was proposed by Day et al. [28] on the basis of single motor unit studies in the hand. His statement was that direct electrical stimulation through the skull preferentially activates corticospinal fibers directly within a few millimeters of the cell body. This is referred to as direct activation and results in D waves (D ¼ direct stimulation of the corticospinal tract at the level of the axonal hillock of the pyramidal cell product in the corticospinal fiber) conducted down the pyramidal system. Further studies [29] revealed that pyramidal neurons could be activated transsynaptically only at higher intensities. In contrast, the lowest threshold form of TMS over the hand area of the motor cortex tends to preferentially activate corticospinal neurons transsynaptically, resulting


in I waves (I = indirect stimulation of the corticospinal tract via cortical interneurons relaying on the pyramidal cell from which the corticospinal fiber is originating) in the pyramidal tract. With higher stimulus intensities, both direct (D wave) and transsynaptic (I wave) activation occur [28], [29]. This D- and I-wave nomenclature represents the heritage of classical neurophysiology in experimental animals in which it has been demonstrated that the former (D waves) derive from direct penetration of the stimulating current in the corticospinal axon, while the latter (I waves) progressively represent later discharges of the same corticospinal axon activated by a progressively higher number of interneurons [30]. The result is that the EMG responses that are recorded around excitability threshold often occur 1–2 ms later than those recorded following TES of the brain [6]. Why there should be this difference between electrical and magnetic forms of stimulation is unclear at the present time but is largely thought to be due to the fact that while TES is mainly eliciting D waves, TMS—at threshold intensities—is mainly eliciting I waves [6]. The response of lower limb muscles has a similar latency with electrical and magnetic stimulation. This suggests that both techniques have the same activation site in the initial segmental or proximal nodes of pyramidal axons as they leave the cortex and readily produce D-wave activity [25]. Safety

Since 1985, many thousand subjects have been examined using low-repetition-rate magnetic stimulators to assess motor function of the peripheral and central nervous systems. There is now a considerable volume of data supporting the safety of magnetic stimulation. There have been no ill effects reported with magnetic stimulation of the peripheral nervous system, and, in the case of cortical stimulation, the incidence of side effects has been very low and within that expected by available statistics for various patient groups. The main area of concern has been the triggering of epileptiform activity in individuals at a high risk for epilepsy. Since TMS has been successfully used in the study of epilepsy and the determination of the site of the epileptic focus, there have only been a few reports of seizures occurring at or shortly after the magnetic stimulation [31]. However, the more recently used rTMS can, depending on the stimulation parameters, evoke seizures in normal subjects and in patients with neurological disease [32]. The presence of pacemakers and other electronic implants is also considered as a contraindication because of damage of the internal electronics due to the induced electric fields and currents resulting from the magnetic pulse. Since implanted metal structures in the brain will have mechanical forces exerted on them due to induced currents, they also should be regarded as a contraindication [4]. Those who are interested in this particular subject are encouraged to read the omnicomprehensive document that was recently settled by a panel of experts and is currently in publication stage [33]. Procedure and Measured Parameters Procedure

Stimulation of the motor cortex is in most cases achieved via a handheld circular coil over the scalp. The subject feels only a moderate tapping sensation on the scalp and the limb twitch. Recordings in humans are made from surface EMG electrodes attached to the skin overlying peripheral muscles using an


89

EMG device. The stimulator triggered the sweep of a standard EMG device, enabling the measurement of the latency between the stimulus and the onset of the response. Apart from the latency, the measurements commonly include the threshold, amplitude, and configuration. Measured Parameters

The threshold reflects the global excitability of the motor pathway and—according to international guidelines—it is defined as the intensity of TMS that produces an identifiable MEP of 50–100 lV in about 50% of ten to 20 consecutive occasions [34]. Threshold in human adults is almost independent of age [35], [36], gender, and hemisphere but varies with different target muscles. A lower threshold is observed in thoracic (51%) than in pelvic (73%) limbs. Even in upper limbs, the threshold is lowest for hand muscles and highest for proximal arm muscles. This may reflect the larger cortical motor areas controlling the hand muscles [26]. Amplitude refers to the recorded voltage of the response. It may be measured from the baseline to the negative peak or from the negative to the positive peak (peak-to-peak amplitude, Figure 5). Usually, amplitude is expressed in absolute terms as micro or millivolts. Sometimes, it is expressed as a percentage of the maximal response after stimulation of the appropriate peripheral nerve. The amplitude can have a high degree of intertrial and intraindividual variability, in particular, during slightly overthreshold stimulating sessions with increasing stimulus intensity or when a subject makes a voluntary effort in a muscle [6]. MEPs are increased and become less variable in their amplitude presumably as a consequence of moving into a flatter region of the stimulus/response curve. The variability appears to be generated spontaneously and may be explained, at least in part, by differences in the state of relaxation of the muscles. As will be

Lat

0

0 0. 00 5 0. 01 0. 01 5 0. 02 0. 02 5 0. 03 0. 03 5 0. 04 0. 04 5 0. 05

Peak-to-Peak

Time (s) Fig. 5. TMS impulse happens at time zero (artifact was partly removed by software). An MEP (shown in the figure) is characterized by its amplitude and latency. Amplitude refers to the recorded voltage of the response and can be measured from the baseline to the negative peak or from the negative to the positive peak (peak-to-peak amplitude is shown in the figure). Latency is the interval between the delivery TMS impulse (time zero) and the resulting response and is expressed in milliseconds (ms).

90


further discussed, on facilitation, even mild muscle contraction increases the amplitude and is therefore difficult to state whether the response has been elicited with the muscle relaxed or not. Also, little alterations in the position of the magnetic stimulating coil over the surface of the cranium can result in wide changes in the variability of MEP responses to TMS, which may reflect fluctuations either in the proportion of available target corticospinal neurones close to threshold or in the magnitude of the induced current under the coil. However, both clamping of the coil relative to the head in studies done by Ellaway et al. [37] and stabilizing the coil position via a mechanic holder failed to reduce the variability of the MEP amplitude. Therefore, some of the observed variability in amplitude is likely to result from spontaneous changes in the size of the descending volley from the cortex, and hence, it reflects the fluctuations in the excitability of the motor cortex already observed. Latency is the interval between the delivery TMS impulse and the resulting response, and it reflects the total motor conduction time from cortex to the target muscle (Figure 4). The latency may be measured at the onset of the motor response and is expressed in milliseconds (ms). MEP latency is affected by the fiber diameter and myelin sheath thickness, as well as by the number of synapses that the impulse must cross. Physiological and clinical studies have focused on the shortest latency responses to provide an estimation of conduction velocities in the fastest descending spinal tracts. Much of the signal is dominated by conduction in a few large fibers. Those fastest descending tracts have been shown to connect monosynaptically to spinal alpha motoneurons [5]. However, there have been reports of longer latency responses being recorded from extensor and flexor carpi radialis muscles of the thoracic limbs and from tibialis anterior and triceps surae muscles of the pelvic limbs in healthy human subjects. In addition to the fastest descending corticospinal fibers, there exist other indirect descending corticospinal pathways with fibers that terminate on the spinal interneurons in the intermediate zone of the spinal gray matter. These descending pathways to spinal motor cells via polysynaptic networks may mediate MEPs with longer latencies. As in experimental animals and also in humans, one single cortical stimulation is able to produce multiple descending volleys along the individual pyramidal fibers [29], [30]. Both spatial and temporal summation of impulses reaching the spinal motoneuron are necessary before it fires; therefore, reduction in the descending volley due to conduction block in some fibers or to loss of the fastest conducting fibers by degeneration and use of slower ones will lead to excitation of a fringe of spinal motoneurons with a longer latency, as it can be inferred by single motor unit recordings [38]. The configuration of the MEPs evoked in the muscles of the hand is in most instances bi- or triphasic. A polyphasic configuration (more than five phases) has to be considered as abnormal in those muscles, whereas a polyphasic configuration is more frequently seen with MEPs evoked in more proximal muscles and muscles of the leg even in normal subjects. In children, however, MEPs are generally polyphasic in early childhood and gradually become triphasic, reaching adult levels at the age of 13 years. Researchers also describe an influence of the stimulation intensity on the configuration of MEPs. At just overthreshold levels, cortical stimulation produces EMG responses that are generally comparable with those following stimulation of peripheral nerves. However, at higher intensities, the cortical


responses (regardless whether after electrical or magnetic stimulation) become polyphasic, because of the multiple descending volleys set up by moderate-to-high levels of cortical stimulation and also to recruitment of larger cortical surface containing many more interneurons. Motor units may fire on receipt of any one of the several excitatory postsynaptic potentials (EPSPs) which these volleys release. This gives rise to an asynchronous activation of motor units in muscle, reducing the maximal amplitude of response and increasing its duration. Conduction times along the central tracts (i.e., pyramidal fibers) governing upper (proximal and distal muscles) and lower limb which can be found in the literature but are around 6 ms for the arm and 12 ms for the leg. Factors Influencing the Latency and Amplitude of MEPs

Slight voluntary contraction of the target muscles shortens the onset latency, and increases the excitability (by lowering TMS threshold) and amplitude [4], [26], [29]. The underlying mechanisms for facilitation are not entirely understood. Some researchers suggest that the effect would be caused by changes in cortical excitability: voluntary contraction increases the size and number of descending volleys evoked by a given stimulus. In the voluntary contraction state, the summation of the descending voluntary impulses from cortical areas, afferent impulses from muscle spindles, and descending potentials secondary to magnetic brain stimulation can activate the spinal motor neurons earlier than under resting conditions, hence shortening the MEP onset latency [6], [7]. Various reports confirmed that the latency shortening can range from 2 to 3.5 ms and that the amplitude during voluntary contraction can be increased to 150–1,500% of that recorded during the resting state. They also showed, by simultaneously recording the evoked responses at the epidural space of the spinal cord and at the muscle, that the shortened latency and increased amplitude of the MEPs during voluntary contraction originated more from changes in spinal than supraspinal (cortical) excitability. Presumably, voluntary activation drives the resting potential of spinal motoneurons close to their postsynaptic action potential threshold, and therefore, the initial descending volley becomes capable of firing at least some motoneurons. This can explain the predominant role of spinal excitability [29]. Response latency shortening during voluntary contraction is likely to rely upon Henneman’s size principle: the first corticomotoneuron cells to fire during a voluntary contraction are those with a smaller diameter and a slower propagation velocity; meanwhile, by increasing the strength of contraction, larger diameter, faster propagating corticospinal and spinal motoneurons are recruited [39], thus shortening the final latency of the muscle response. Facilitation can be exploited when muscle responses at rest are absent even with maximal TMS intensity or too small to be reliably measured because of central nervous system pathology [e.g., spinal cord injury, multiple sclerosis (MS), stroke, etc.]. However, in this situation, the noise of the ongoing muscle activity and the small amplitude and variability of MEPs (due to loss of synchronization in spinal motoneuronal firing) may make precise measurement of MEPs onset latency arguable. Therefore, complete relaxation of the muscle but moderate contraction of the same muscle on the opposite site (contralateral facilitation) would result in the same degree of latency reduction and amplitude increase [40].


Effect of Coil Position

Commonly, the lowest threshold for stimulation occurs when induced current in the brain (which flows in a direction opposite to that flowing in the coil) flows from the posterior to anterior at an angle approximately perpendicular to the line of the central sulcus. As a general rule, when the coil is placed over the scalp, maximal amplitude and shortest latency responses are elicited on the right if the stimulating current in the coil flows counterclockwise. For the largest and shortest response on the left, the current flow must be reversed. The reasons that underlie this effect are poorly understood but presumably reflect differences in the direction and distribution of current flow within the brain produced by the two orientations of the coil. However, during voluntary contraction, about 20% of the MEPs were maximum (amplitudes) on the left when the current flowed counterclockwise and vice versa. The most stable responses were obtained from muscles of the thoracic or pelvic limb, respectively, with the round coil centered on the vertex or slightly anterior to the vertex. For a given stimulus intensity, responses are always largest in distal muscles, particularly, in hand muscles. Effect of Age, Height, and Gender

The latency time is strongly correlated with height. No statistical difference was reported between gender for threshold and amplitude. The statistical difference found between gender for latency seems to result from the correlation between latency and height [4]. It is concluded that height is an important parameter in defining the MEP limits. Most studies however have concentrated on the 20–50-year range, and there has been no comprehensive description of the normative results of hand and leg MEPs over a wide range of ages in a substantial sample of male and female subjects. Some authors described a linear increase in latency with increasing age, but with a relatively weak or no correlation at all, while a statistically significant increase of the intensity of TMS to reach excitability threshold with aging has been noted [35], [36]. Both height and age had a significant effect on the leg MEP latencies. An interesting finding is the different effect that was seen regarding age on the hand and leg MEP latency. In the cortical motor area, 75% or more of Betz cells showed age-related morphological changes, while changes of small pyramidal neurons were less severe than those of Betz cells. Anthropomorphic parameters are therefore important in defining normal MEPs, especially in the lower limbs. In young children, latency of the MEP, however, does not attain adult values until about 11 years. Several other parameters are described in the literature but less employed in routine research and clinical activity; they include the silent period (that is the inability of the subject to produce any EMG activity for some time—whose duration depends on the intensity of TMS and not on muscle contraction—immediately following a MEP), motor maps (that is the extension on the scalp of representation of a given muscle), and interhemispheric differences of all the measured parameters. All these information can be vastly found in the literature (see [41] for a review). Indications and Results Diagnosis

A diagnostic test should be sufficiently accurate and should have specified clinical indications. The high reproducibility in standardized conditions makes MEPs efficacious in evaluating


91

Latency is the interval between the delivery TMS impulse and the resulting response and reflects the total motor conduction time from cortex to the target muscle.

corticospinal tract functionality even in subclinical involvement of central motor pathways (see [42] for a review). In patients with spinal cord disorders, MEPs may be useful in demonstrating the site of spinal cord lesion and also to monitor the disease, as for example, in cervical spondylotic myelopathy. Serial MEP recording might be useful in ascertaining progressive forms and for selection of patients who may benefit from surgical treatment. Next to its application in spinal cord lesions, the technique is also extremely useful in MS. Several reports [5] suggest that the technique is more sensitive than other types of evoked potentials in MS. Presently, electrophysiology is the only means of confirming suspected amyotrophic lateral sclerosis (ALS) [43]. In addition to EMG performed to determine lower motor neuron involvement, the introduction of TMS has

allowed the assessment of central motor pathway function. A common finding is that TMS fails to evoke a muscle response or evokes a response with a reduced amplitude despite high intensity stimulation. The MEP latency is only modestly prolonged. This can be attributed by the degeneration of the corticomotor cells or reduced firing frequency in corticospinal fibers with consequent impaired temporal summation at the motoneuron. Moreover, it provides a sensitive means for the assessment and monitoring of upper motor neuron involvement in motor neuron disease (see [42] for a review). Prognosis

Potential (µV)

The quality of motor recovery after stroke is difficult to predict on the basis of only clinical data. Since 1989, studies have been conducted to assess the value of MEPs in patients with stroke [44]. The applicao Fp1 Fp2 Real 45 TMS tion of TMS has yielded o TMS Sham 45 contradictory results, probF7 F8 ably because of the great variF3 F4 ability of patients included and differences in the methodologies used. However, most F3 C3 Cz C4 T4 authors agree that the evoked potentials measured in the acute stage had a predictive Pz value. For instance, the 45o P4 P3 absence of responses to TMS T5 T6 in the first 48 h is predictive of absent or very poor func–20 O1 O2 tional hand motor recovery. The use of TMS as a predicµV 0 0 300 tive test in patients with 10 ma traumatic cervical spinal cord TMS injury does not provide more useful information regarding –30 Cz N10 Real 45°TMS motor recovery than the Sham 45°TMS physical examination but may be of benefit in uncooperative N40 N100 or incomprehensive patients. N10 0 Follow-up of stroke patients P14 P100 P80 may provide fruitful informaP30 tion for final outcome (see [45] for a review). +30 TMS is a valuable diagnos100 200 –100 0 300 tic tool for detection of lesions Time (ms) TMS along the spinal cord but found no linear correlation Fig. 6. TMS-EEG responses during stimulation of the C3 scalp position of the 10–20 internabetween the clinical motor tional nomenclature for EEG recording sites. Notice the different peaks elicited by the stimustatus and the electrophysiolus ( TMS = 0) starting from about 10 ms (used with permission from [52]). logical changes. For facial 92



nerve outcome after acoustic neuroma surgery, no prognostic significance of preoperative TMS was reported. Monitoring

There are two clinical indications for monitoring neurological function during surgical procedures: to detect inadvertent damage early when the resulting dysfunction might still be reversible and to guide the surgeon with regard to the extent of safe operative resection (e.g., in tumor resection) or curve correction (e.g., in scoliosis surgery). For these purposes, the ideal neurophysiological technique should 1) have high sensitivity and specificity 2) provide real-time feedback, 3) not intrude physically into the operative field, 4) not hinder access for the anesthetist, 5) not prolong the operation unduly, 6) not be subject to artifactual changes that could be misinterpreted as incipient or actual neural dysfunction (i.e., there should be a low incidence of false positives), and 7) be equally useful in patients with and without preexisting neurological deficits. Since the introduction of TMS, MEPs as well as SEPs are recorded routinely in many centers, for more complete information on both the descending corticomotoneuron tracts and ascending sensory pathways during major spinal surgery. However, in animal experiments, motor pathways have been reported to be more susceptible than sensory pathways to spinal cord trauma and ischemia; therefore, identifiable motor dysfunction might commonly be expected to precede sensory dysfunction, and a technique to monitor conduction in the corticospinal pathways (e.g., TMS) is advisable. In addition to spinal cord monitoring during spinal operations, other potential roles for the technique may be sciatic nerve function during hip replacement, peripheral nerve, and brachial plexus integrity during hand surgery, facial nerve function during surgery, and spinal cord status during aortic surgery. Therapy

TMS is able to modify neuronal activity locally and at distant sites when delivered in a series of trains of pulses (rTMS). Data from stimulation of the motor cortex suggest that the type of effect on the excitability of the cortical network depends on the frequency of stimulation. Studies with rTMS for the treatment of psychiatric disorders (depression, schizophrenia, [46], [47]) and neurological disorders (Parkinson disease, taskrelated dystoniatic disorders, and epilepsy) were promising for the future, and the authors concluded that the technique may possess tremendous potential as a treatment for these disorders [48]. However, much research is still needed to optimize the technical considerations, such as stimulus frequency, intensity, and magnetic coil position, and to investigate the neurophysiological changes. Conclusions

Since the pioneering works on motor conduction, several other application areas of TMS have emerged. Along with rapidrate TMS (repetition rates of up to 60 pulses/s), coils designed for focal stimulation and image-guided targeting of stimulation to desired cortical structures, neuronal processes can now be disrupted even in relatively well-defined cortical areas. Delivering two sequential pulses with independently adjusted stimulus intensities and with a short ISI (1–200 ms) to the primary motor cortex (M1) allowed the exploration of inhibition and facilitation within the motor pathway (paired-pulse TMS; [49], [50]). In an attempt to assess the extent and loci of


Fig. 7. TMS-MRI integrated representation. Color grid contains topography of eight different upper limb muscles from which MEPs were simultaneously recorded during stimulation of individual scalp sites (used with permission from [69]).

local and remote TMS-induced brain effects, which was not possible with TMS and the EMG alone, functional brain imaging was combined with TMS. Reflecting changes in cerebral blood flow and oxygenation, 15O–H2O or 18F–FDG positron emission tomography, 99 mTc ethylcysteinate dimer singlephoton emission tomography, blood-oxygenation-leveldependent magnetic resonance imaging (MRI), and nearinfrared spectroscopy showed bilateral cortical activity as well as activation of subcortical structures and the cerebellum. The introduction of TMS-compatible EEG allowed one to measure the instant and direct neuronal effects of TMS [51], [52] (Figure 6); bilateral activation patterns, similar to functional imaging, were exhibited in source images derived from the EEG. Intriguing examples of the use of TMS/EEG technique in studying corticocortical connectivity are already available in the literature [53]. TMS shows a great promise for future clinical applications. Presently, the key topics in TMS research include altered cortical excitability in neurological diseases [42], [54], [55], functional relevance of cortical areas in cognitive task performance, and treatment of psychiatric diseases. To date, however, there are not enough data to establish TMS studies as part of clinical diagnostics or therapy in any neurological or psychiatric disease [56]. Recently, patterned types of stimulation have been introduced, each still under scrutiny, to define its efficacy in research and clinical settings as well as its potential hazards (see [33]). The last decade has seen a rapid increase in the applications of TMS to study cognition, brain–behavior relations, and the pathophysiology of various neurologic and psychiatric disorders [41], [46], [48], [56]–[68]. In addition, evidence has accumulated that demonstrates that TMS provides a valuable tool for interventional neurophysiology applications, modulating brain activity in a specific, distributed, corticosubcortical network. Finally, recent evidences have shown that multiple muscle recordings allow to probe dynamic properties of the multiple representations at the cortical levels to orchestrate the best movement performance [69] (Figure 7). Paolo Maria Rossini received his honor’s degree in medicine from the Catholic University of Rome in July 1974. From 1974 to 1978, he attended the postgraduate course as a


93

resident in neurology. He was appointed as assistant professor in 1978 and subsequently transferred to the University of Chieti, where he remained until 1984. During this period, he took part in various courses and research projects in the United States, collaborating with Roger Cracco and his group on somatosensory-evoked potentials. He was appointed as associate professor in 1985 and transferred to the University of Rome Tor Vergata, where he taught clinical neurophysiology. From 1984 to 1990, he actively participated in the development of clinical applications in TMS and TES, collaborating with many Italian and foreign colleagues. He worked as a visiting professor at the University of Irvine (California) in 1989. In 1990, he was appointed chair of the Department of Neurology at Fatebenefratelli Hospital San Giovanni Calibita of Rome. From 1995 to 2003, he was the editor-in-chief of Electroencephalography and Clinical Neurophysiology (later Clinical Neurophysiology). He worked as a scientific director of the Scientific Institute on Dementia and Psychiatry (recognized by the Italian Ministry of Health) between 1997 and 2007. In 2000, he was appointed as full professor of clinical neurology at the Faculty of Medicine at the University of Campus Biomedico in Rome. Since 2006, he has been the scientific director of Integrated Research Centre University Campus Bio-Medico of Rome. His current research fields include brain plasticity in healthy adult brain and following brain lesions, TMS in cognitive sciences, neurophysiology of aging brain and of dementias, and integrated brain functional imaging. Luca Rossini received his master’s degree in biomedical engineering from the University Campus Bio-Medico of Rome (Italy) in 2006. He is currently a Ph.D. student in biomedical engineering at the laboratories of Biomedical Robotics and Biomicrosystems of the University Campus Bio-Medico of Rome. He currently works as a research fellow at the University Campus BioMedico of Rome, Italy. He is the first editor of Brain Machine Interfaces for Space Applications: Enhancing Astronauts’ Capabilities. His current research interests are on noninvasive neural interfaces for the control of assistive technologies and biomechatronic prosthesis and human movement decomposition into discrete-controlled submovements. His research interests include space applications of noninvasive neural interfaces and on liquid ventilation for water-immersed astronauts to enhance acceleration sustainability. He is a Member of the IEEE and the IEEE Engineering in Medicine and Biology Society. Florinda Ferreri received her honor’s degree in medicine in 2001 and a certification in neurology in 2006 from the University Campus Bio Medico of Rome, Italy, under the direct supervision of Prof. Rossini. She is currently a locum neurologist consultant with expertise in neurophysiology, teaching, and research duties at the University Hospital Campus Biomedico of Rome and Ph.D. student in clinical neurophysiology at the University 94


of Kuopio, Finland. Her current research fields include TMS in cognitive sciences, integrated brain functional imaging (particularly EEG-TMS), neurophysiology of aging brain, and dementias. She is a member of Italian Society of Clinical Neurophysiology and Italian Society of Neurology from 2002. Address for Correspondence: Paolo Maria Rossini, M.D., University Campus Bio-Medico, Rome, Via Alvaro del Portillo 21, 00122 Rome, Italy. E-mail: [email protected]. References [1] P. A. Merton and H. B. Morton, ‘‘Stimulation of the cerebral cortex in the human subject,’’ Nature, vol. 285, p. 227, 1980. [2] P. A. Merton, H. B. Morton, D. K. Hill, and C. D. Marsden, ‘‘Scope of a technique for electrical stimulation of human brain, spinal cord and muscle,’’ Lancet, vol. 2, pp. 597–600, 1982. [3] A. T. Barker, R. Jalinous, and I. L. Freeston, ‘‘Non-invasive stimulation of human motor cortex,’’ Lancet, vol. 1, pp. 1106–1107, 1985. [4] A. T. Barker, I. L. Freeston, R. Jalinous, and J. A. Jarratt, ‘‘Magnetic stimulation of the human brain and peripheral nervous system: An introduction and the results of an initial clinical evaluation,’’ Neurosurgery, vol. 20, pp. 100–109, 1987. [5] P. M. Rossini, M. G. Marciani, M. Caramia, V. Roma, and F. Zarola, ‘‘Nervous propagation along ‘‘central’’ motor pathways in intact man: Characteristics of motor responses to ‘‘bifocal’’ and ‘‘unifocal’’ spine and scalp non-invasive stimulation,’’ Electroencephalogr. Clin. Neurophysiol., vol. 61, pp. 272–286, 1985. [6] P. M. Rossini, M. Caramia, and F. Zarola, ‘‘Mechanisms of nervous propagation along central motor pathways: Noninvasive evaluation in healthy subjects and in patients with neurological disease,’’ Neurosurgery, vol. 20, pp. 183–191, 1987. [7] P. M. Rossini, M. Caramia, and F. Zarola, ‘‘Central motor tract propagation in man: Studies with non-invasive, focal, scalp stimulation,’’ Brain Res., vol. 415, pp. 211–225, 1987. [8] S. Rossi, S. F. Cappa, C. Babiloni, P. Pasqualetti, C. Miniussi, F. Carducci, F. Babiloni, and P. M. Rossini, ‘‘Prefrontal cortex in long-term memory: An «interference« approach using magnetic stimulation,’’ Nat. Neurosci., vol. 4, pp. 948–952, 2001. [9] R. Mariorenzi, F. Zarola, M. D. Caramia, C. Paradiso, and P. M. Rossini, ‘‘Non-invasive evaluation of central motor tract excitability changes following peripheral nerve stimulation in healthy humans,’’ Electroencephalogr. Clin. Neurophysiol., vol. 81, pp. 243–250, 1991. [10] K. Stefan, E. Kunesch, L. G. Cohen, R. Benecke, and J. Classen, ‘‘Induction of plasticity in the human motor cortex by paired associative stimulation,’’ Brain, vol. 123, pp. 572–584, 2000. [11] J. Classen, B. Steinfelder, J. Liepert, K. Stefan, P. Celnik, L. G. Cohen, A. Hess, E. Kunesch, R. Chen, R. Benecke, and M. Hallett, ‘‘Cutaneomotor integration in humans is somatotopically organized at various levels of the nervous system and is task dependent,’’ Exp. Brain Res., vol. 130, p. 4859, 2000. [12] V. Di Lazzaro, A. Oliviero, P. Profice, E. Saturno, F. Pilato, A. Insola, P. Mazzone, P. Tonali, and J. C. Rothwell, ‘‘Muscarinic receptor blockade has differential effects on the excitability of intracortical circuits in the human motor cortex,’’ Exp. Brain Res., vol. 135, pp. 455–461, 2000. [13] Y. Z. Huang, M. J. Edwards, E. Rounis, K. P. Bhatia, and J. C. Rothwell, ‘‘Theta burst stimulation of the human motor cortex,’’ Neuron, vol. 45, pp. 201– 206, 2005. [14] V. Di Lazzaro, F. Pilato, M. Dileone, P. Profice, A. Oliviero, P. Mazzone, A. Insola, F. Ranieri, M. Meglio, P. A. Tonali, and J. C. Rothwell, ‘‘The physiological basis of the effects of intermittent theta burst stimulation of the human motor cortex,’’ J. Physiol., vol. 586, pp. 4481–4487, 2008. [15] R. Gentner, K. Wankerl, C. Reinsberger, D. Zeller, and J. Classen, ‘‘Depression of human corticospinal excitability induced by magnetic theta-burst stimulation: Evidence of rapid polarity-reversing metaplasticity,’’ Cereb. Cortex, vol. 19, pp. 2056–2053, 2008. [16] E. Iezzi, A. Conte, A. Suppa, R. Agostino, L. Dinapoli, A. Scontrini, and A. Berardelli, ‘‘Phasic voluntary movements reverse the aftereffects of subsequent theta-burst stimulation in humans,’’ J. Neurophysiol., vol. 100, pp. 2070–2076, 2008. [17] M. Hamada, Y. Terao, R. Hanajima, Y. Shirota, S. Nakatani-Enomoto, T. Furubayashi, H. Matsumoto, and Y. Ugawa, ‘‘Bidirectional long-term motor cortical plasticity and metaplasticity induced by quadripulse transcranial magnetic stimulation,’’ J. Physiol., vol. 586, pp. 3927–3947, 2008. [18] R. Plonsey, Biomagnetic Phenomena. New York: McGraw-Hill, 1969. [19] F. Grandori and P. Ravazzani, ‘‘Magnetic stimulation of the motor cortex— Theoretical considerations,’’ IEEE Trans. Biomed. Eng., vol. 38, pp. 180–191, 1991. [20] B. J. Roth, L. G. Cohen, M. Hallett, W. Friauf, and P. J. Basser, ‘‘A theoretical calculation of the electric field induced by magnetic stimulation of a peripheral nerve,’’ Muscle Nerve, vol. 13, pp. 734–741, 1990. [21] L. Heller and D. B. van Hulsteyn, ‘‘Brain stimulation using electromagnetic sources: Theoretical aspects,’’ Biophys. J., vol. 63, pp. 129–138, 1992. [22] H. Nakamura, H. Kitagawa, Y. Kawaguchi, and H. Tsuji, ‘‘Direct and indirect activation of human corticospinal neurons by transcranial magnetic and electrical stimulation,’’ Neurosci. Lett., vol. 210, pp. 45–48, 1996.


[23] T. Kriggs, B. R. Buchbinder, W. E. Butler, K. H. Chiappa, H. J. Jiang, B. R. Rosen, and G. R. Cosgrove, ‘‘Stereotactic transcranial magnetic stimulation: Correlation with direct electrical cortical stimulation,’’ Neurosurgery, vol. 41, pp. 1319–1325, 1997. [24] A. T. Barker, C. W. Garnham, and I. L. Freeston, ‘‘Magnetic nerve stimulation: The effect of waveform on efficiency, determination of neural membrane time constants and the measurement of stimulator output,’’ Electroencephalogr. Clin. Neurophysiol., vol. 43, pp. 227–237, 1991. [25] J. C. Rothwell, ‘‘Techniques and mechanisms of action of transcranial stimulation of the human motor cortex,’’ J. Neurosci. Methods, vol. 74, pp. 113–122, 1997. [26] J. C. Rothwell, B. L. Day, P. D. Thompson, J. P. Dick, and C. D. Marsden, ‘‘Some experiences of techniques for stimulation of the human cerebral motor cortex through the scalp,’’ Neurosurgery, vol. 20, pp. 156–163, 1987. [27] P. S. Tofts, ‘‘The distribution of induced currents in magnetic stimulation of the brain,’’ Phys. Med. Biol., vol. 35, pp. 1119–1128, 1990. [28] B. L. Day, D. Dressler, A. Maertens de Noordhout, C. D. Marsden, K. Nakashima, J. C. Rothwell, and P. D. Thompson, ‘‘Electric and magnetic stimulation of human motor cortex: Surface EMG and single motor unit responses,’’ J. Physiol., vol. 412, pp. 449–473, 1989. [29] V. Di Lazzaro, D. Restuccia, A. Oliviero, P. Profice, L. Ferrara, A. Insola, P. Mazzone, P. Tonali, and J. C. Rothwell, ‘‘Magnetic transcranial stimulation at intensities below active motor threshold activates intracortical inhibitory circuits,’’ Exp. Brain Res., vol. 119, pp. 265–268, 1998. [30] H. D. Patton and V. E. Amassian, ‘‘Single and multiple-unit analysis of cortical stage of pyramidal tract activation,’’ J. Neurophysiol., vol. 17, pp. 345–363, 1954. [31] R. Cantello, S Rossi, C. Varrasi, M. Ulivelli, C. Civardi, S. Bartalini, G. Vatti, M. Cincotta, A. Borgheresi, G. Zaccara, A. Quartarone, D. Crupi, A. Lagana`, M. Inghilleri, A. T. Giallonardo, A. Berardelli, L. Pacifici, F. Ferreri, M. Tombini, F. Gilio, P. Quarato, A. Conte, P. Manganotti, L. G. Bongiovanni, F. Monaco, D. Ferrante, and P. M. Rossini, ‘‘Slow repetitive TMS for drug-resistant epilepsy: Clinical and EEG findings of a placebo-controlled trial,’’ Epilepsia, vol. 48, pp. 366–374, 2007. [32] E. M. Wassermannn, ‘‘Risk and safety of repetitive transcranial magnetic stimulation: Report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5–7, 1996,’’ Electroencephalogr. Clin. Neurophysiol., vol. 108, pp. 1–16, 1998. [33] S. Rossi, M. Hallett, P. M. Rossini, and A. Pascual-Leone, and the Safety of TMS Consensus Group‘‘Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. A consensus statement from the International Workshop on Present and Future of TMS: Safety and Ethical Guidelines, Siena, March 7–9, 2008,’’ Clin. Neurophysiol., to be published. [34] P. M. Rossini, A. T. Baker, A. Berardelli, M. D. Caramia, G. Caruso, R. Q. Cracco, M. R. Dimitrijevic, M. Hallet, Y. Katayama, C. H. Lu¨cking, et al., ‘‘Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: Basic principles and procedures for routine clinical applications. Report of an IFCN committee,’’ Electroencephalogr. Clin. Neurophysiol., vol. 91, no. 2, pp. 79–92, 1994. [35] P. M. Rossini, M. T. Desiato, and M. D. Caramia, ‘‘Age-related changes of motor evoked potentials in healthy humans: Non-invasive evaluation of central and peripheral motor tracts conductivity and excitability,’’ Brain Res., vol. 593, pp. 14–19, 1992. [36] P. M. Rossini, S. Rossi, C. Babiloni, and J. Polich, ‘‘Clinical neurophysiology of aging brain: From normal aging to neurodegeneration,’’ Prog. Neurobiol., vol. 83, pp. 375–400, 2007. [37] P. H. Ellaway, N. J. Davey, D. W. Maskill, S. R. Rawlinson, H. S. Lewis, and N. P. Anissimova, ‘‘Variability in the amplitude of skeletal muscle responses to magnetic stimulation of the motor cortex in man,’’ Electroencephalogr. Clin. Neurophysiol., vol. 109, pp. 104–113, 1998. [38] P. M. Rossini, M. D. Caramia, C. Iani, M. T. Desiato, G. Sciarretta, and G. Bernardi, ‘‘Magnetic transcranial stimulation in healthy humans: Influence on the behavior of upper limb motor units,’’ Brain Res., vol. 676, pp. 314–324, 1995. [39] E. Hennemann, G. Somjen, and D. G. Carpenter, ‘‘Functional significance of cell sizes in spinal motoneurons,’’ J. Neurophysiol., vol. 28, pp. 560–580, 1965. [40] C. W. Hess, K. R. Mills, and N. M. F. Murray, ‘‘Responses in small hand muscles from magnetic stimulation of the human brain,’’ J. Physiol., vol. 388, pp. 397–420, 1987. [41] P. M. Rossini and S. Rossi, ‘‘Transcranial magnetic stimulation: Diagnostic, therapeutic and research potential,’’ Neurology, vol. 68, pp. 484–488, 2007. [42] P. M. Rossini and S. Rossi, ‘‘Clinical applications of motor evoked potentials,’’ Electroencephalogr. Clin Neurophysiol., vol. 106, pp. 180–194, 1998. [43] U. Ziemann and A. Eisen, ‘‘TMS for ALS: Why and why not,’’ Clin Neurophysiol., vol. 115, no. 6, pp. 1479–1485, 2004. [44] P. Cicinelli, R. Traversa, and P. M. Rossini, ‘‘Post-stroke reorganization of brain motor output to the hand: A 2–4 months follow-up with focal magnetic transcranial stimulation,’’ Electroencephalogr. Clin. Neurophysiol., vol. 105, pp. 438–450, 1997.


[45] P. M. Rossini, G. Calautti, F. Pauri, and J. C. Baron, ‘‘Post-stroke plastic reorganization in the adult brain,’’ Lancet Neurol., vol. 2, pp. 493–502, 2003. [46] M. S. George, E. M. Wassermann, W. A. Williams, J. Steppel, A. PascualLeone, P. Basser, M. Hallett, and R. M. Post, ‘‘Changes in mood and hormone levels after rapid-rate transcranial magnetic stimulation (rTMS) of the prefrontal cortex,’’ J. Neuropsychiatry Clin. Neurosci., vol. 8, pp. 172–180, 1996. [47] C. Miniussi, C. Bonato, S. Bignotti, A. Gazzoli, M. Gennarelli, P. Pasqualetti, G. B. Tura, M. Ventriglia, and P. M. Rossini, ‘‘Repetitive transcranial magnetic stimulation (rTMS) at high and low frequency: An efficacious therapy for major drug-resistant depression?’’ Clin. Neurophysiol., vol. 116, pp. 1062–1071, 2005. [48] E. M. Wassermann and S. H. Lisanby, ‘‘Therapeutic application of repetitive transcranial magnetic stimulation: A review,’’ Clin. Neurophysiol., vol. 112, pp. 1367–1377, 2001. [49] T. Kujirai, M. Sato, J. C. Rothwell, and L. G. Cohen, ‘‘The effect of transcranial magnetic stimulation on median nerve somatosensory evoked potentials,’’ Electroencephalogr. Clin. Neurophysiol., vol. 89, pp. 227–234, 1993. [50] F. Ferreri, G. Curcio, P. Pasqualetti, L. De Gennaro, R. Fini, and P. M. Rossini, ‘‘Mobile phone emissions and human brain excitability,’’ Ann. Neurol., vol. 60, pp. 188–196, 2006. [51] R. J. Ilmoniemi, J. Virtanen, J. Ruohonen, J. Karhu, H. J. Aronen, R. Na¨a¨ta¨nen, and T. Katila, ‘‘Neuronal responses to magnetic stimulation reveal cortical reactivity and connectivity,’’ Neuroreport, vol. 8, pp. 3537–3540, 1997. [52] C. Bonato, C. Miniussi, and P. M. Rossini, ‘‘Transcranial magnetic stimulation and cortical evoked potentials: A TMS/EEG co-registration study,’’ Clin. Neurophysiol., vol. 117, pp. 1699–1707, 2006. [53] M. Massimini, F. Ferrarelli, R. Huber, S. K. Esser, H. Singh, and G. Tononi, ‘‘Breakdown of cortical effective connectivity during sleep,’’ Science, vol. 309, pp. 2228–2232, 2005. [54] D. L. Gilbert, M. A. Garvey, A. S. Bansal, T. Lipps, J. Zhang, and E. M. Wassermannn, ‘‘Should transcranial magnetic stimulation research in children be considered minimal risk?’’ Clin. Neurophysiol., vol. 115, pp. 1730–1739, 2004. [55] A. A. Ku¨hn, S. A. Brandt, A. Kupsch, T. Trottenberg, J. Brocke, K. Irlbacher K, G. H. Schneider, and B. U. Meyer, ‘‘Comparison of motor effects following subcortical electrical stimulation through electrodes in the globus pallidus internus and cortical transcranial magnetic stimulation,’’ Exp. Brain Res., vol. 155, pp. 48–55, 2004. [56] M. Kobayashi and A. Pascual-Leone, ‘‘Transcranial magnetic stimulation in neurology,’’ Lancet Neurol., vol. 2, pp. 145–156, 2003. [57] A. A. Gershon, P. N. Dannon, and L Grunhaus, ‘‘Transcranial magnetic stimulation in the treatment of depression,’’ Amer. J. Psychiatry, vol. 160, pp. 835– 845, 2003. [58] C. A. Tassinari, M. Cincotta, G. Zaccara, and R. Michelucci, ‘‘Transcranial magnetic stimulation and epilepsy,’’ Clin. Neurophysiol., vol. 114, pp. 777–798, 2003. [59] S. Rossi and P. M. Rossini, ‘‘TMS in cognitive plasticity and the potential for rehabilitation,’’ Trends Cogn. Sci., vol. 8, pp. 273–279, 2004. [60] J. P. Lefaucheur, X. Drouot, F. Von Raison, I. Me´nard-Lefaucheur, P. Cesaro, and J. P. Nguyen, ‘‘Improvement of motor performance and modulation of cortical excitability by repetitive transcranial magnetic stimulation of the motor cortex in Parkinson’s disease,’’ Clin. Neurophysiol., vol. 115, pp. 2530– 2541, 2004. [61] R. E. Hoffman, R. Gueorguieva, K. A. Hawkins, M. Varanko, N. N. Boutros, Y. T. Wu, K. Carroll, and J. H. Krystal, ‘‘Temporoparietal transcranial magnetic stimulation for auditory hallucinations: Safety, efficacy and moderators in a fifty patient sample,’’ Biol. Psychiatry, vol. 58, pp. 97–104, 2005. [62] J. L. Couturier, ‘‘Efficacy of rapid-rate repetitive transcranial magnetic stimulation in the treatment of depression: A systematic review and meta-analysis,’’ J. Psychiatry Neurosci., vol. 30, pp. 83–90, 2005. [63] F. Fregni, D. K. Simon, A. Wu, and A. Pascual-Leone, ‘‘Non-invasive brain stimulation for Parkinson’s disease: A systematic review and meta-analysis of the literature,’’ J. Neurol. Neurosurg. Psychiatry, vol. 76, pp. 1614–1623, 2005. [64] M. Hallett, ‘‘Transcranial magnetic stimulation: A primer,’’ Neuron, vol. 55, pp. 187–199, 2007. [65] M. S. George, Z. Nahas, J. J. Borckardt, B. Anderson, M. J. Foust, C. Burns, S. Kose, and E. B. Short, ‘‘Brain stimulation for the treatment of psychiatric disorders,’’ Curr. Opin. Psychiatry, vol. 20, pp. 250–254, 2007. [66] J. Ma´lly and T. W. Stone, ‘‘New advances in the rehabilitation of CNS diseases applying rTMS,’’ Expert Rev. Neurother., vol. 7, pp. 165–177, 2007. [67] J. T. Devlin and K. E. Watkins, ‘‘Stimulating language: Insights from TMS,’’ Brain, vol. 130, pp. 610–622, 2007. [68] M. C. Ridding and J. C. Rothwell, ‘‘Is there a future for therapeutic use of transcranial magnetic stimulation?’’ Nat. Rev. Neurosci., vol. 8, pp. 559–567, 2007. [69] J. M. Melgari, P. Pasqualetti, F. Pauri, and P. M. Rossini, ‘‘Muscles in concert: Study of primary motor cortex upper limb functional topography,’’ PLoS One, vol. 3, no. 8, p. e3069, 2008.


95

CLASSIFIED EMPLOYMENT To conform to the Age Discrimination in Employment Act, and to discourage age discrimination, IEEE may reject any advertisement containing any of these phrases, or similar ones: ‘‘recent college grad,’’ ‘‘1-4 years maximum experience,’’ ‘‘up to 5 years experience,’’ or ‘‘10 years maximum experience.’’ IEEE reserves the right to append to any advertisement, without specific notices to the advertiser, ‘‘Experience ranges are suggested minimum requirements, not maximums.’’ IEEE assumes that, since advertisers have been notified of this policy in advance, they agree that any experience requirements, whether stated as ranges or otherwise, will be construed by the reader as minimum requirements only. While IEEE does not ban the use of the term ‘‘entry level’’ its use is discouraged since, to some, it connotes an age rather than experience designation. To place an ad in the IEEE Engineering in Medicine and Biology Magazine Employment Opportunities section, call: Susan E. Schneiderman Business Development Manager þ1 732 562 3946

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE 96

